Fault detection and classification of an HVDC transmission line using a heterogenous multi‐machine learning algorithm

This paper presents a novel integrated multi-Machine Learning (ML) system architecture for the protection of bipolar HVDC transmission line in which different ML models Support Vector Machine (SVM) and K-Nearest Neighbours (KNN) are used fault detection classification. The KNN type classifier is designed as dual-purpose module, not only detects but also acts redundant module unsure declaration from startup unit. Gradients standard deviations DC current, voltage, harmonic correlation coefficient between aerial zero modes current appropriate feature vector extracted single-end signal measurement. Overall, 154 training cases 53 main test obtained by simulating various non-fault states on ±650 kV-1000 km Current Source Converter (CSC)–HVDC using an EMTDC/PSCAD platform. modules trained MATLAB tested under severe conditions with total 2220 cases. Thanks to proposed architecture, results show that algorithm effective enough detect distinguish variety internal faults pseudo-faults/external faults. Also, it needs low data requirements.